Click to enlarge

On this day a year ago, the UK started its piloting of mass testing in the city of Liverpool using lateral flow tests. The latest graphic in the #ChemVsCOVID series with the Royal Society of Chemistry looks at how these tests work, and how accurate are they compared to the standard PCR tests that are usually used for testing.

The lateral flow tests used for rapid testing for COVID are similar in function to home pregnancy tests. Just as pregnancy tests use antibodies to test for the presence of a specific hormone, COVID lateral flow tests use antibodies to detect the presence of the SARS-CoV-2 virus.

After you’ve gagged and winced your way through the swabbing process, you pop it into a buffer solution. If you’re infected, virus particles pass into the solution from the swab. The solution is then added to the test device.

What happens next is in the name: the sample flows up the test device. The plastic casing of the device houses a nitrocellulose strip, which the sample flows along. This isn’t too dissimilar from the rudimentary paper chromatography experiments you might have carried out in school chemistry classes, where water carries pen inks along paper.

As the sample flows up the strip, it passes through different zones. The first of these is the conjugate pad, sometimes referred to as the reaction zone. Here, mobile antibodies which bind to parts of the virus particles are present. If virus particles are present in the sample, these antibodies latch on to them and piggyback their way further up the test strip.

The next part of the test strip is the test zone. Here, another set of antibodies that can bind to parts of the virus are present. Unlike the first set of antibodies, these are stuck in place, and when they grab the virus particles they’re stuck in a pincer grip between the two antibodies. The mobile antibodies also have gold nanoparticles attached, and when they get stuck in place in the test zone they cause a red line to appear.

Not all of the mobile antibodies from the reaction zone will grab hold of virus particles, but they get dragged up the test strip by the flow regardless. The control zone contains fixed antibodies that can bind directly to the unbound mobile antibodies, with the gold nanoparticles again causing a red line to appear. This confirms that the test has worked correctly.

So, a sample from an infected person should cause red lines to appear in both the test zone and the control zone. If a person isn’t infected, they should only see a red line in the control zone. If a red line doesn’t appear in the control zone, the test hasn’t worked correctly, so whether there’s a red line in the test zone or not is irrelevant.

Up until the point at which lateral flow tests were more widely introduced, PCR tests were the main testing method used to diagnose COVID. Inevitably, once lateral flow tests started to be used, there were questions about how accurate they might be.

To answer the question of accuracy, we need to break it down into two terms: sensitivity and specificity.

Sensitivity is a measure of the correct production of positive test results. In other words, it gauges how many patients who have the virus correctly test positive on a lateral flow test. Exact figures can vary depending on the brand of test used, but for the Innova tests used in the UK Public Health England has estimated a figure of 76.8%. This means that, for roughly every four people that are infected and should test positive, one tests negative: a false negative.

By comparison, the range for sensitivity for PCR tests is estimated to be between 85-98% by a range of studies. At the lower end of the range, this would mean 3 in 20 people who should test positive would return a negative result – at the higher end, it would be 1 in 50.

It’s clear, then, that lateral flow tests are more likely to return false negative results, which could mean infected patients thinking they’re all clear. This is one of the reasons why regular repeated testing is recommended with lateral flow tests, as this can partly overcome this limitation. It’s thought that false negatives are more likely with lateral flow tests if you have a lower viral load – that is, fewer virus particles in your blood.

Another component of tests’ accuracy is specificity, a measure of the correct production of negative results. In other words, it tells us how many patients who are not infected with the virus correctly test negative on a test. For lateral flow tests, the specificity is 99.7%. This would mean that, for every 1000 patients who should test negative, only three would return a positive result – a false positive. This compares pretty well to PCR tests, which have a specificity even closer to 100%.

What does all this mean? Well, it means that, if you test positive with a lateral flow test, you can be pretty confident it’s not a false positive – though confirmatory testing with a PCR test helps remove any doubt. If you test negative, it’s still probable that you don’t have the virus, but you can be even more confident if you’ve tested negative on multiple lateral flow tests.

With that said, there was some criticism of the lateral flow tests used in the Liverpool trial in the UK, with one study stating that they detected just 49% of COVID infections in asymptomatic people. Subsequent studies based on the figures from the pilot have also found lower sensitivity in practice than the 76.8% figure determined by Public Health England.

While the concerns over missed infections are valid, it remains the case that lateral flow tests have enabled much easier detection of asymptomatic cases, particularly in countries like the UK where symptoms are a prerequisite of access to PCR testing. Allowing more infected people to isolate is key to reducing the spread of COVID. In addition, some countries have limited capacity for PCR testing, so the availability of these rapid tests helps support their efforts.

References/further reading